Back to skill

Security audit

DAF Intent Guard — 意图漂移检测

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local Python helper for deciding whether a changed user request should patch, replan, or abort an ongoing task, with no evidence of hidden access or unsafe behavior.

Install only if you want an agent helper that influences when ongoing multi-step tasks should continue, roll back, or stop. Be aware that its README is mainly Chinese and its built-in semantic rules are simple heuristics, so users should treat its decisions as guidance rather than authoritative task control.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill description and all usage guidance are presented entirely in Chinese, with no indication that other languages are supported or that the locale restriction is intentional and justified. This can violate a language/locale policy when users are not given an opt-in or alternative language option.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The date parser explicitly recognizes Chinese-only natural-language terms such as "今天", "明天", and "后天" alongside some English terms, which creates language-specific behavior in the skill. There is no surrounding comment, argument, or documentation indicating user opt-in for this locale behavior or that the skill is intentionally region-specific.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.